Accurate prediction of peptide-HLA class I binding is crucial for immunotherapy and vaccine development, but existing methods often struggle to capture the intricate biological relationships between peptides and diverse HLA alleles. Here, we introduce SeqDA-HLA, a pan-specific prediction model that combines language model-based embeddings (ELMo) with a dual attention mechanism-self-aligned cross-attention and self-attention-to capture rich contextual features and pairwise interactions. Evaluations against 14 state-of-the-art methods on multiple benchmark datasets demonstrate that SeqDA-HLA consistently outperforms competing approaches, achieving an AUC value up to 0.9856 and accuracy as high as 0.9408. Notably, SeqDA-HLA maintains robust performance across peptide lengths (8-14) and HLA alleles, showcasing its generalizability. Beyond predictive accuracy, SeqDA-HLA offers interpretability by highlighting essential anchor residues and revealing key binding motifs, thereby aligning with experimentally validated biological insights. As a further demonstration of practical impact, we fine-tune SeqDA-HLA on an Influenza virus dataset, successfully predicting binding changes induced by single amino acid mutations. Overall, SeqDA-HLA serves as a powerful and interpretable tool for peptide-HLA binding prediction, with potential applications in epitope-based vaccine design and precision immunotherapy.

MOTIVATION: High-throughput phenomic projects generate complex data from small treatment and large control groups that increase the power of the analyses but introduce variation over time. A method is needed to utlize a set of temporally local controls that maximizes analytic power while minimizing noise from unspecified environmental factors. RESULTS: Here we introduce 'soft windowing', a methodological approach that selects a window of time that includes the most appropriate controls for analysis. Using phenotype data from the International Mouse Phenotyping Consortium (IMPC), adaptive windows were applied such that control data collected proximally to mutants were assigned the maximal weight, while data collected earlier or later had less weight. We applied this method to IMPC data and compared the results with those obtained from a standard non-windowed approach. Validation was performed using a resampling approach in which we demonstrate a 10% reduction of false positives from 2.5 million analyses. We applied the method to our production analysis pipeline that establishes genotype-phenotype associations by comparing mutant versus control data. We report an increase of 30% in significant P-values, as well as linkage to 106 versus 99 disease models via phenotype overlap with the soft-windowed and non-windowed approaches, respectively, from a set of 2082 mutant mouse lines. Our method is generalizable and can benefit large-scale human phenomic projects such as the UK Biobank and the All of Us resources. AVAILABILITY AND IMPLEMENTATION: The method is freely available in the R package SmoothWin, available on CRAN http://CRAN.R-project.org/package=SmoothWin. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.